SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

CANCER

1Center for Systems Biology, Massachusetts General Hospital, 185 CambridgeStreet, CPZN 5206, Boston, MA 02114, USA. 2Department of Radiology, Massachu-setts General Hospital, 185 Cambridge Street, CPZN 5206, Boston, MA 02114, USA.3Department of Systems Biology, Harvard Medical School, 200 Longwood Avenue,Boston, MA 02115, USA. 4Graduate Program in Immunology, Harvard MedicalSchool, Boston, MA 02115, USA. 5Center for Immunology and Infectious Disease,Massachusetts General Hospital, 149 8th Street, Charlestown, MA 02129, USA.6Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard MedicalSchool, Boston, MA 02115, USA.*These authors contributed equally to this work.†Corresponding author. Email: mpittet@mgh.harvard.edu

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

2017 © The Authors,

some rights reserved;

exclusive licensee

American Association

for the Advancement

of Science.

htD

ownloaded from

In vivo imaging reveals a tumor-associatedmacrophage–mediated resistance pathway inanti–PD-1 therapySean P. Arlauckas,1,2,3* Christopher S. Garris,1,4* Rainer H. Kohler,1 Maya Kitaoka,5

Michael F. Cuccarese,1 Katherine S. Yang,1 Miles A. Miller,1,2 Jonathan C. Carlson,1

Gordon J. Freeman,6 Robert M. Anthony,5 Ralph Weissleder,1,2,3 Mikael J. Pittet1,2†

Monoclonal antibodies (mAbs) targeting the immune checkpoint anti–programmed cell death protein 1 (aPD-1) havedemonstrated impressive benefits for the treatment of some cancers; however, these drugs are not always effective,andwestill havea limitedunderstandingof themechanisms that contribute to their efficacyor lack thereof.Weused invivoimaging to uncover the fate and activity of aPD-1mAbs in real time and at subcellular resolution inmice.We show thataPD-1mAbs effectively bind PD-1+ tumor-infiltrating CD8+ T cells at early time points after administration. However, thisengagement is transient, and aPD-1mAbs are capturedwithinminutes from the T cell surfacebyPD-1− tumor-associatedmacrophages. We further show that macrophage accrual of aPD-1mAbs depends both on the drug’s Fc domain glycanandon Fcg receptors (FcgRs) expressedby hostmyeloid cells and extend these findings to the human setting. Finally, wedemonstrate that in vivo blockadeof FcgRs before aPD-1mAbadministration substantially prolongs aPD-1mAbbindingto tumor-infiltrating CD8+ T cells and enhances immunotherapy-induced tumor regression inmice. These investigationsyield insight into aPD-1 target engagement in vivo and identify specific Fc/FcgR interactions that can be modulated toimprove checkpoint blockade therapy.

tp://

by guest on A

ugust 27, 2020stm

.sciencemag.org/

INTRODUCTIONImmune checkpoint blockade is a recent development in cancer therapythat has shown remarkable results in certain cancers and patient groups(1–3). Currently approved immune checkpoint blockers aremonoclonalantibodies (mAbs) that target the programmed cell deathprotein 1 (PD-1)or cytotoxic T lymphocyte–associated protein 4 pathways, and agentstargeting other pathways are in clinical development (including OX40,Tim-3, and LAG-3) (4). Checkpoint inhibitors are used to reactivateexhausted tumor-specific T cells and reinstate cancer immunosurveillance(5, 6). Some cancer tissues limit antitumor immunity by up-regulatingimmunosuppressive factors such as PD-1 ligand (PD-L1) that binds toPD-1 on tumor-specific CD8+T cells (7). Drugs targeting the PD-1/PD-L1immune checkpoint axis can block immunosuppressive signals and en-able T cell–mediated elimination of cancer cells (8). However, immunecheckpoint blockade is not always effective, andwe lack a complete under-standing of the mechanisms that contribute to efficacy and resistance (9).

At present, experimental and clinical evidence suggests that a pre-existing tumor infiltrate of CD8+ T cells is one of the most favorableprognostic indicators of checkpoint inhibitor response (10). Also, pa-tients with the highest degree of neoantigen burden (high mutationalload) in their cancers may have increased tumor infiltration by T cellsand more robust responses to checkpoint blockade (11, 12). Histologyand sequencing methodologies have been used to define metrics of cy-totoxic T cell infiltration in tumors (13), with a focus on the identifica-

tion of tumor neoantigens and the resultant antigen-specific T cellexpansion after immunotherapy (11). These studies have providedinsight into the mechanism of anti–PD-1 (aPD-1) mAb–induced anti-tumor T cell activation and spurred efforts focused on identifying newstrategies that foster T cell recruitment to tumors (14–16).

Much less is known about checkpoint inhibitors’ in vivo pharmaco-kinetics and interactionswith host components in the tumor bed. Studyingthese parameters is likely essential to identifying resistancemechanismsand developing improved therapeutic options. Here, we used intravitalimaging to follow fluorescently labeled aPD-1 mAbs in real time and atsubcellular resolution. Because tumor microenvironments are home todiverse host cell types, and immune checkpoint blockers are unlikely tosolely act on T cells, we focused on aPD-1mAb interactions with varioushost components by simultaneously assessing aPD-1 mAbs, tumor cells,CD8 T cells, and myeloid cells/macrophages. We investigated tumor-infiltratingCD8+T cells because they express PD-1 and are the expectedtargets of aPD-1mAbs.We also investigated myeloid cells because theyare frequently found in the stroma of growing tumors (17), and emergingevidence indicates that they can affect virtually all therapeutic modalities,including immunotherapy (18). Our results not only confirm existingknowledge onPD-1 inhibition but also uncover findingswith therapeuticimplications to further improve immunotherapy.

RESULTSGlobal aPD-1 mAb biodistributionWe initially sought to track the temporal distribution of aPD-1mAbsin vivo at the organ level. We thus covalently labeled aPD-1 mAb(clone 29F.1A12) with an Alexa Fluor 647 dye (AF647–aPD-1) usingN-hydroxysuccinimide (NHS) chemistry. For maximal brightnesswithout dye quenching, we optimized the labeling conditions to achieveabout four fluorochrome molecules per antibody. For in vitro studies,the EL4mouse lymphoma cell line was used as a T cellmodel because ofits stable PD-1 expression and its broad adaptation to in vitro culture

1 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

http://stm.sciencem

ag.orgD

ownloaded from

(19). Using this cell line, we confirmed that fluorescent labeling ofaPD-1 did not interfere with the drug’s binding specificity (Fig. 1A).In addition, AF647–aPD-1 retained therapeutic activity in the ovalbumin-expressing MC38 tumor model, which is responsive to single-agentaPD-1 therapy (Fig. 1B) (20). Collectively, these data indicate thatAF647–aPD-1 retains PD-1 tropism and antitumor activities.

We next examined the in vivo biodistribution of AF647–aPD-1 inthe wild-typeMC38 tumormodel, which also responds to aPD-1 treat-ment, but less efficiently than the ovalbumin-expressingMC38 counter-part inwhich aPD-1 treatment results in uniform tumor rejection.Mousecohorts were sacrificed at times ranging from 0.5 to 72 hours after treat-ment, and organs were removed for fluorescencemeasurements (Fig. 1C).AF647–aPD-1 signal was primarily retainedwithin tumors over time (Fig.1D and fig. S1). We observed a large spike of AF647–aPD-1 in the liver,lungs, kidney, and spleen at 0.5 hours after injection, followed by a sub-sequent decrease over time that coincided with increases in AF647–aPD-1 signal in the tumor (Fig. 1D). Collectively, these data indicatethat the drug started to accumulate in tumors within minutes afterinjection but that maximal aPD-1 accumulation in the tumor wasachieved after 24 hours.

Cellular kinetics and dynamics of aPD-1After observing that aPD-1 mAbs collect within the tumor micro-environment shortly after administration, we further aimed to studywhether the drugs bind their intended target T cells at the tumor site.To this end, we used intravital microscopy in dorsal skinfold chambers,which enabled us to examine the distribution and tropism of AF647–aPD-1 at subcellular resolutionwithin the tumor stromaand longitudinallyafter drug administration (Fig. 2A). The experimental system allowedsimultaneous tracking of four components: aPD-1 mAbs (labeled withAF647), MC38 tumor cells (labeled withH2B-mApple), T cells [labeled

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

with green fluorescent protein (GFP) or yellow fluorescent protein(YFP)], and tumor-associatedmacrophages (TAMs) (labeledwith PacificBlue–dextran nanoparticles). We used two different reporter mousemodels to visualize T cells: (i) DPE-GFP mice (21) in which all GFP-expressing cells areCD90+ (fig. S2A) and (ii) interferon-g (IFN-g) reporter(GREAT) mice (22), which were useful because tumor-infiltrating YFP+

cells in these mice were almost exclusively CD8+ T cells (fig. S2B). Thefluorescent Pacific Blue–dextran nanoparticle has been validated for in-tratumoralmacrophage identification (23), andwe confirmed its specificityfor macrophages (F4/80+ cells) in the tumor stroma (fig. S2C). Finally,we verified that single-agent aPD-1 treatment was able to suppressMC38–H2B-mApple tumor growth in the window chamber system (Fig. 2B).

Upon administration of AF647–aPD-1, we found that the antibodyrapidly perfused tumor vessels and gradually disseminated out of thevasculature and into the tumor interstitium (Fig. 2C and movie S1).AF647–aPD-1was observed onGFP-labeledT cells as early as 5min afterinjection, and these were the first cells in the tumormicroenvironment tobe detectably labeled by the drug. AF647–aPD-1 binding to tumor-infiltrating T cells was initially pericellular but, within minutes, formedpuncta on the cell surface (fig. S3). These rearrangements occurredwithout apparent decreased T cell motility (fig. S4, A to C). Later timepoints revealed that TAMs, which were stationary in the tumor micro-environment (fig. S4D), had collectedmost of theAF647–aPD-1; T cellswere not associated with AF647–aPD-1 at these time points (Fig. 2, Dand E). T cells were present in the tumormicroenvironment at all timesexamined, removing the possibility that the drug biodistribution was anartifact of T cell loss after therapy. Also, tracingAF647–aPD-1 across alltime points failed to show binding to tumor cells, precluding the possi-bility that aPD-1 mAbs had direct effects on cancer cells (fig. S5).

Quantification of the tumor microenvironment images taken at15 min or 24 hours after drug administration showed a pattern of

by guest on August 27, 2020

/

Fig. 1. aPD-1mAb labeling facilitates tracking of tissue biodistribution. (A) The rat anti-mouse PD-1 29F.1A12 clone, conjugated to AF647 via NHS ester linkage, efficientlybinds PD-1+ T cells (here, EL4 cells) as detected by flow cytometry (gray histogram). Isotype control staining is shown inwhite. (B) MC38 tumors were equally responsive to single-dose AF647–aPD-1 and unlabeled aPD-1, whereas tumor sizes increased 72 hours after control immunoglobulin G2a (IgG2a) treatment. (C) Fluorescence reflectance imaging ofthree tumors compared to tumor-draining (tdLN) and nondraining (ndLN) lymphnodes 24 hours after AF647–aPD-1 treatment (AF647: lex = 620 to 650 nm, lem= 680 to 710 nm).Scale bar, 5 mm. (D) Quantified AF647–aPD-1 in each tissue demonstrating tumor accumulation. Values represent SEM and n = 3 unless otherwise noted. ***P < 0.001, unpaired,two-tailed t test.

2 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on August 27, 2020

http://stm.sciencem

ag.org/D

ownloaded from

AF647 signal loss on T cells over time and a concomitant AF647 signalincrease on macrophages (Fig. 3A). Although DPE-GFP labels CD90+

lymphocytes (fig. S2A), the GREAT mouse model allowed us to focusspecifically on IFN-g–expressing cells (movie S2), which are almostexclusively CD8+ T cells in the tumor microenvironment (fig. S2B).Intravital imaging of these cells confirmed surface binding of AF647–aPD-1 within minutes after drug administration (fig. S3). Longitudinalstudies further showed that T cells eventually lose AF647–aPD-1mAbs,which are physically transferred to, and retained by, neighboring macro-phages (Fig. 3B). Consistent with our observations inDPE-GFPmice, wefound in IFN-g reporter mice that aPD-1 transfer from T cells to macro-phages limited the overall duration of drug binding to their intendedtarget cells (Fig. 3C).

To address whether checkpoint blockade agent uptake by macro-phages could be independently validated using a bulk tissuemeasurement,we performed flow cytometry of tissues excised from tumor-bearing

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

animals at 0.5 and 24 hours after AF647–aPD-1 administration (Fig. 3D).We evaluated several relevant cell populations for aPD-1 binding andconfirmed our intravital microscopy observations: aPD-1 mAbs wereboundmostly toCD8+T cells at 0.5 hours but tomacrophages at 24 hours(Fig. 3,DandE). Tumormacrophageswere positive for rat IgG2a (fig. S6),confirming that the aPD-1mAb, and not just the fluorophore, was trans-ferred in vivo.Other cell types investigated, includingCD4+Tcells, CD45+

CD11b+ F4/80− cells (which can include granulocytes and monocytes),anddendritic cells, didnotdisplay significant bindingof aPD-1 at any timepoint tested (Fig. 3, D and E).

Analysis of B16 melanoma and KP1.9 lung adenocarcinoma tumormodels also demonstrated that most AF647–aPD-1 mAbs accumulatewithin TAMs at 24 hours, similar to the findings obtained with MC38colon adenocarcinoma (fig. S7A).Most of the intratumoralmacrophageswere bound byAF647–aPD-1 in all three tumormodels (fig. S7B). Linearregression analysis of combined data further indicated that aPD-1 mAb

Fig. 2. In vivo temporal aPD-1mAbpharmacokinetics reveals drug accumulation in TAMs. (A) Diagramdepicting intravital imaging setupwith labeled aPD-1,MC38 tumorcells, T cells, and TAMs. FMX-PacBlue, ferumoxytol–Pacific Blue. (B) Treatmentwith single-dose aPD-1mAb can achieve remission in theMC38–H2B-mApple tumormodel. Tumorsare outlined in gray. Scale bars, 2mm. (C) Z-projections of anMC38–H2B-mApple tumor in a DPE-GFPmouse injected intravenously with AF647–aPD-1 after 15min (top) or24 hours (bottom). (D) Micrographs of T cells (magenta outline) identified as GFP+ cells and TAMs (yellow outline) identified by Pacific Blue signal. Outlines are overlaid onmicrographs of the corresponding AF647–aPD-1 channel. Scale bars, 30 mm. (E) Intravital microscopy biodistribution studies indicate early aPD-1 binding to T cells and long-term accumulation in TAMs. Data are representative of five independently treated DPE-GFP mice and normalized to autofluorescent signal.

3 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on August 27, 2020

http://stm.sciencem

ag.org/D

ownloaded from

uptake by macrophages is independent of macrophage number and canalso occur when TAM numbers are relatively low (fig. S7C).

aPD-1 mAb transfer mechanismTo explore the removal of aPD-1 mAbs on T cells by macrophages, wefirst asked whether the latter may also express PD-1 on their surface.However, ex vivo flow cytometry analysis indicated that TAMs, in con-trast to tumor-infiltratingCD8+T cells, were PD-1− (Fig. 4A). PD-1wasalso absent in natural killer andB cells in these tumors (fig. S8).We thendesigned an in vitro coculture system combining bonemarrow–derivedmacrophages and T cells that constitutively express PD-1 to create acontrolled system to study themechanism of drug collection bymacro-phages (Fig. 4B). T cells were preincubatedwithAF647–aPD-1 to emulatethe antibody initially bound to T cells we observed in vivo and then co-cultured with macrophages. Using this experimental setting, AF647–aPD-1 mAbs effectively relocated from T cells to macrophages withinseveral minutes, as detected by the formation of drug puncta withinmacrophages (Fig. 4, B and C, and movie S3). The transfer could not beattributed solely to phagocytosis of cell debris because it occurred even inthe presence of the phagocytosis inhibitor dynasore (Fig. 4B). No macro-

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

phage uptake was observed when using an AF647-labeled isotype controlIgG antibody, which did not bind T cells (fig. S9).

Because tumor macrophages did not capture AF647–aPD-1 in sub-stantial quantities early after administration, did not express PD-1 ontheir cell surface, and withdrew drug bound to the T cell surface, wereasoned that AF647–aPD-1 mAbs must accumulate in macrophagesthrough a non–antigen-specific mechanism. The aPD-1 clone 29F.1A12is a rat IgG2a isotype that is used to mimic the biological properties ofhuman IgG4. Both rat IgG2a and human IgG4 have been demonstratedto bind inhibitory FcgRs (mouse FcgRIIb and human FcgRIIB, respec-tively) (24, 25). Adding FcgRIIb/III blocking antibodies to the in vitrococulture system diminished AF647–aPD-1 transfer from T cells tomacrophages (Fig. 4, B and C). The effect was specific because blockingFcgRIV, an Fc receptor that does not bind rat IgG2a, failed to inhibitAF647–aPD-1 transfer (Fig. 4, B and C).

To substantiate these findings, we developed a flow cytometry–based antibody transfer assay:Macrophages were incubated with T cellspreviously labeled with AF647–aPD-1 mAbs for 30 min and analyzedby flow cytometry. AF647–aPD-1 signal was detected on macrophagesin this experimental setting (Fig. 4D); however, aPD-1 transfer could be

Fig. 3. In vivo imaging reveals aPD-1 mAb transfer from CD8+ T cells to TAMs. (A) Quantified AF647 signal on T cells and TAMs from a representative DPE-GFP mousedemonstrates collection of AF647–aPD-1 in T cells at 15 min and in TAMs at 24 hours after injection of therapy. (B) Early time-course intravital microscopy images acquired in anIFN-g–YFP reporter mouse with an MC38–H2B-mApple tumor. Yellow arrows indicate the site of aPD-1 mAb binding to an IFN-g+ CD8 T cell at 6 to 15 min and macrophageinternalization at times >21min. Scale bar, 30 mm. (C) Quantification of aPD-1mAb on IFN-g–expressing CD8+ lymphocytes and TAMs reveals a narrowwindow of target binding.(D) Flow cytometry histograms pregated for 7-AAD− (7-aminoactinomycin D–negative)/CD45+ showAF647–aPD-1 signal (x axis, logarithmic scale) on immune cell populations at0.5 and 24 hours after administration (blue). Cell populations from untreated control animals (red) were used as reference. DC, dendritic cell; Granul./Mono., granulocyte/monocyte.(E) aPD-1 mAb binds to CD8+ lymphocytes early but accumulates in TAMs at later time points. **P < 0.01; ****P < 0.0001, unpaired, two-tailed t test.

4 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on August 27, 2020

http://stm.sciencem

ag.org/D

ownloaded from

neutralized by adding a blocking antibody to FcgRIIb/III to the co-culture system (Fig. 4, D and E). AF647–aPD-1mAbs added directly tothe culturemedium in the absence of T cells were not efficiently taken up

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

by macrophages, further suggesting thatT cells are the major source of aPD-1 formacrophages (Fig. 4, D and E). We alsotested whether aPD-1 loss on the T cellsurface might be due to receptor internal-ization independentofmacrophageuptake.T cells were exposed to AF647–aPD-1 for1 hour at 37°C, treated with an acid solu-tion to remove cell surface antibodies, andanalyzed by flow cytometry to assess re-maining (internalized) fluorescent signals(fig. S10A) (26). Acid stripping stronglyreducedAF647–aPD-1 detection, indicat-ing that antibody internalization is likelynot the primary contributor to aPD-1 losson T cells (fig. S10B).

Macrophages coculturedwithAF647–aPD-1–coated T cells were positive notonly for AF647 but also for rat IgG2a, asassessed by enzyme-linked immunosorbentassay (fig. S11A). The eventual decline inmacrophage rat IgG2a was not accompa-nied by the release of IgG2a into the super-natant, suggesting that acquired antibodyis eventually degraded by the macrophage(fig. S11B). Collectively, these data indicatethat aPD-1 mAb removal from the T cellsurface receptor by macrophages is apharmacologic end point elicited by FcgRinteractions with T cell–bound antibodycomplexes.

To assess whether the PD-1 receptoris transferred during aPD-1 removal,PD-1+ T cells were exposed to unlabeledaPD-1, cocultured with macrophages (toenable aPD-1 capture), and reprobed forcell surface PD-1 expression with a fluo-rescent aPD-1 mAb. Transfer of unlabeledaPD-1 together with PD-1 would pre-vent PD-1 detection upon reprobing withfluorescent aPD-1. Instead, T cells fromwhich aPD-1 had been captured were effi-ciently reprobed, indicating that aPD-1 re-moval frees up PD-1 molecules, whichbecome available to fluorescent aPD-1mAb(fig. S12A). Control experiments confirmedthat the increased reprobing signal was notcontributed by new PD-1 molecules onthe surface of T cells (fig. S12B). Together,these data indicate that PD-1 remains onthe T cell membrane after aPD-1 capture.

Because Fc/FcgR binding interactionsofmany IgG subclasses dependuponmAbFc glycosylation (27), we profiled the glycanstructures from the murine aPD-1 mAb(29F.1A12) and further extended our anal-

ysis to the human aPD-1 mAb nivolumab. Both antibodies were treatedwith peptide N-glycosidase F (PNGase F), a glycosidase that cleaves N-linked glycan, to remove glycan from each antibody, and the digested

Fig. 4. aPD-1 mAb transfer to macrophages is mediated by FcgRs. (A) Ex vivo flow cytometry histograms of MC38tumors stainedwith phycoerythrin (PE)–aPD-1 show that CD8+ T cells but not TAMs express cell surface PD-1. (B) Cocultureof bonemarrow–derivedmacrophages (Mø) and AF647–aPD-1–coated EL4 lymphocytes was used to quantify the AF647–aPD-1 puncta in macrophages preblocked with Fcg receptor (FcgR) inhibitors or the phagocytosis inhibitor dynasore. *P <0.05; ****P < 0.0001, one-way analysis of variance (ANOVA). n.s., not significant. (C) Representative images of macrophageslabeled with PKH red and observed by time-lapse imaging for aPD-1mAb (yellow) transfer from neighboring lymphocytes(cerulean) after 30 min. FcgRII/III inhibition blocked antibody transfer (insets). Scale bars, 10 mm. (D and E) Flow cytometrywas used to estimate AF647–aPD-1 transfer to F4/80+ bone marrow–derived macrophages in coculture assays when themAb was added directly (green), bound to EL4 cells (red), or bound to EL4 cells in the presence of FcgRII/III neutralizingantibody (blue). Results in (D) are a representative histogram of AF647 signal in macrophages, and (E) is the geometricmean fluorescence intensity (G-MFI) value of conditions presented in (D). Data are from three independent experiments. *P <0.05; ** P < 0.01, two-way ANOVA with Tukey’s multiple comparisons test. Values represent SEM for three separateexperiments.

5 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

http://stm.scienc

Dow

nloaded from

products were analyzed by high-performance liquid chromatography(HPLC) (Fig. 5A).We found that murine and human aPD-1 mAbs sharethe same predominant glycoform that lacks terminal galactose residues(G0F) and is fucosylated on the penultimate N-acetylglucosamine (fig.S13). Both mouse and human aPD-1 contained substantial fractions ofterminally galactosylated glycoforms, indicating a high degree of Fc gly-can heterogeneity in these aPD-1 mAb preparations.

Thesimilarity inglycanpatternbetweenmouseandhumanaPD-1mAbsled us to hypothesize that FcgR-mediated antibody transfer is relevantto human aPD-1 interactions.We fluorescently labeled the anti-humanPD-1 mAb nivolumab with AF647 and adapted the in vitro coculturesystem (Fig. 4B) toprimaryhuman cells.Wedifferentiatedhumanmacro-phages from blood monocytes using macrophage colony-stimulatingfactor, and PD-1–expressing CD8+ T cells were generated by isolatingprimary human CD8+ T cells and stimulating them with plate-boundaCD3mAbs for 3 days. Coculture of AF647-nivolumab–labeled humanCD8+ T cells with macrophages resulted in mAb transfer from CD8+

T cells to macrophages (movie S4), and the transfer was inhibited byblocking FcgRs using aFcgRIIB/III (Fig. 5B). There was no evidenceof CD8+ T cell membrane components in the macrophages, consistentwith the antibody alone being removed from the surface PD-1 receptor(fig. S14). Furthermore, blocking FcgRs decreased AF647-nivolumabpuncta inside of macrophages, implying that FcgRs also regulate nivo-lumab uptake in human cells (Fig. 5C).

Improving immunotherapyIn an attempt tominimize aPD-1/FcgR interactions, we used PNGase Fto remove the glycan frommurine aPD-1mAbs and confirmed cleavage

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

of glycan by lens culinaris agglutinin lectin blot (fig. S15A). The PNGaseF–treated aPD-1 was labeled with AF647, and flow cytometry confirmedthat the presence of glycan was not required for aPD-1 tropism becausePD-1+ T cells were still efficiently labeledwith PNGase F–treatedAF647–aPD-1 (fig. S15B). However, live cell imaging demonstrated that glycanremoval diminished antibody transfer from T cells to macrophages (Fig.6A), and these resultswere confirmedby flow cytometry (Fig. 6, B andC).

We then aimed to uncover the in vivo activity of aPD-1mAbs whenaPD-1/FcgR interactions were therapeutically inhibited. To this end, wetracked aPD-1, CD8+ T cells, macrophages, and tumor cells in mice inwhich we inhibited FcgRs by infusing FcgRIIb/III blocking antibodies(2.4G2 clone) before delivering AF647–aPD-1 mAbs (Fig. 6D). Wefound that administration of the FcgR blocking agent substantially pro-longed the occupancy time of AF647–aPD-1 mAbs on CD8+ T cells inthe tumor bed (Fig. 6E). Furthermore, whereas the response of MC38tumors to aPD-1 therapy typically varies among animals (fig. S16),blocking FcgR interactions completely eliminated the fraction ofnonresponders observed, with complete tumor rejection in all micethat received the combination treatment (Fig. 6F). These data provideevidence that mAb/FcgR interactions abbreviate aPD-1 mAb occupancytime on tumor-infiltrating CD8+ T cells and limit therapy response; con-versely, aPD-1 mAb therapy can be improved by blocking FcgR interac-tions (fig. S17).

DISCUSSIONMany cancer patients do not respond to immune checkpoint blockadetherapy, and we lack a complete understanding of the mechanisms that

by guest on August 27, 2020

emag.org/

Fig. 5. Nivolumab shares similar glycan patterns with mouse aPD-1 and is transferred to macrophages via FcgRs. (A) HPLC analysis of the glycan patterns found onthemouse aPD-1mAbandnivolumab shows theG0F (“2”) isoform tobepredominant, but glycosylation is not uniform. (B) AF647-labelednivolumab (yellow)wasused to stain thesurfaceof aCD3-stimulated PKHgreen–labeledhumanCD8+ T cells (cerulean) co-incubatedwith PKH red–labeledperipheral bloodmononuclear cell–derivedmacrophages in thepresence or absence of Fc Block. Scale bars, 20 mm. (C) Quantification of AF647+ puncta permacrophage confirms that nivolumab is transferred via FcgRs. Values represent SEM forfour separate experiments.

6 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

hD

ownloaded from

contribute to treatment efficacy and resistance. Here, by using time-lapse intravital microscopy, we uncovered in real time how the immunecheckpoint blocker aPD-1 mAb distributes in tumors and physicallyinteracts with tumor microenvironment components. This approachenabled us to detect aPD-1 mAb association with cytotoxic T cells in-filtrating tumors in vivo. Furthermore, by following the drug’s pharma-cokinetics over time, we found the drug to be rapidly removed fromPD-1+ CD8+ T cells and transferred to neighboring PD-1− TAMs. Thetransfer of aPD-1 mAbs from T cells to macrophages was unexpectedbecause macrophages do not directly take up aPD-1 mAbs in culture.We further identified that aPD-1 uptake bymacrophages depends bothon the Fc domain of the antibody and on FcgRs expressed by macro-phages. Interactions between the drugs and macrophages are likely im-portant because blocking Fc/FcgR binding inhibited aPD-1 transferfrom CD8+ T cells to macrophages in vivo and enhanced aPD-1 thera-peutic efficacy.

Although clinical aPD-1 has an extensive circulation half-life(~26 days), our observations suggest that the time of target engagementin the local tumor environmentmay bemuch shorter. This engagementtime is reduced at least in part by FcgRII/III, whichmediate aPD-1mAbuptake from T cells to macrophages. Accordingly, previous work inmice has shown that IgG2a isotypes preferentially bind FcgRIIb/IIIand that aPD-1 therapywith an IgG2amAb ismore effective in FcgRIIb

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

knockout animals (25). Our studies further suggest that FcgR-mediatedaPD-1 removal does not involve transfer of cell membrane componentsor trogocytosis, which has been described for other mAbs includingrituximab (27, 28). However, we found that aPD-1 uptake by macro-phages is favored when the mAb is bound to PD-1 on T cells, whichaligns with previous findings that FcgRs bind IgGs more efficientlywhen they form immunecomplexes (29,30).Wealso found thatPD-1 re-mains on the T cell surface after aPD-1 removal, but we did not observethese PD-1molecules to bind new aPD-1. Overall, aPD-1 transfer fromT cells tomacrophages appears to be faster than aPD-1 uptake byT cellsin the tumor stoma.

The human aPD-1 drugs nivolumab and pembrolizumab were de-signed as human IgG4 antibody isotypes that are not known to fixcomplement or trigger antibody-dependent cell-mediated cytotoxicity(ADCC) (31). However, IgG4 can bind FcgRI and FcgRIIb, and theseinteractions can have profound clinical consequences (32, 33). Awarenessof the FcgR binding profile of an mAb offers an opportunity to improveupon existingmonoclonal therapies, exemplified by obinutuzumab, a de-fucosylated IgG1 biosimilar of rituximab designed to bind FcgRIIIA andenhance ADCC against CD20+ cells in chronic lymphocytic leukemia(34). Human germline variants of FcgRs that display altered Fc bindingtropism have been identified and are an important focus in the effort tounderstand responses tomAb therapies that rely onFcgRs for therapeutic

by guest on August 27, 2020

ttp://stm.sciencem

ag.org/

Fig. 6. Disrupting Fc bindingaffectsmacrophageuptakeof aPD-1 and improves treatment efficacy. (A) PKHgreen–labeled EL4 cells (cerulean) stainedwith nativeAF647–aPD-1or deglycosylatedAF647–aPD-1 coculturedwith PKH red–labeledbonemarrow–derivedmacrophages (Mø). Images are representative of three separate experiments. Scalebars, 10 mm. (B) Fluorescence-activated cell sorting plots of mouse Mø (gated on F4/80+) cocultured with PD-1+ EL4 cells labeled with either AF647–aPD-1 mAb (blue) or degly-cosylated AF647–aPD-1mAb (red). (C) aPD-1mAbdeglycosylation substantially reduces the transfer fromEL4 cells toMø (n= 3). **P< 0.01, unpaired two-tailed t test. (D) Intravitalimages 10 and 120 min after AF647–aPD-1 drug extravasation in a representative IFN-g–YFP reporter mouse MC38–H2B-mApple tumor pretreated with the FcgRII/III blockingantibody (2.4G2). (E) Quantified AF647–aPD-1 from three 2.4G2-treated mice shows prolonged aPD-1 binding to tumor T cells, relative to mice treated with AF647–aPD-1 alone(data repeated from Fig. 3C for comparison). (F) MC38 tumor growth curves of mice (n ≥ 5) treatedwith isotype control (black), aPD-1 (blue), and aPD-1 plus 2.4G2 (red). *P < 0.05,one-way ANOVA with Tukey’s multiple comparison test.

7 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on August 27, 2020

http://stm.sciencem

ag.org/D

ownloaded from

function, like cetuximab (35), rituximab (36), trastuzumab (37), and othermAb therapies (38).

We suggest that nivolumab and other IgG4-based mAbs are not ex-ceptions to the rules of FcgR binding, and therefore, Fc interactionsshould be considered in pharmacologic models. This is particularly im-portant because there is growing interest in immune checkpoint mole-cules as diagnostic tools to identify PD-1/PD-L1+ tumors. For example,previous efforts to image PD-1 expression using positron emission to-mography (PET) radioligands have focused on gross tissue distributionand lack the resolution to identify cellular tropisms in vivo.Natarajan et al.used hamster anti-mouse PD-1 (39), but it is not fully understood howhamster mAbs interact with mouse FcgRs in this context. A secondaryaPD-1 PET imaging study reported the use of the RMP1-14 rat IgG2aanti-mouse PD-1 clone to cross-correlatewith ex vivo PD-1 staining (40),but drug withdrawal bymacrophages could complicate the relationshipbetween PET signal and PD-1 expression. Future preclinical diagnosticefforts to image PD-1 expression should consider imaging agents thatavoid FcgR interactions; however, antibodies meant tomimic nivolumaband pembrolizumab should accurately represent the Fc status of thehuman IgG4 antibodies.

The findings from the present study also have potential implicationsfor improving current treatment options. Fromour understanding of Fcinteractions with aPD-1, Fc-engineered IgG variants that abrogate FcgRbinding andmAb effector functions (41) or combinations with therapiesthat inhibit FcgRbinding in vivomay enhance the effects of treatment. Inaddition, therapies designed to target tumor macrophages (18), whencombined with aPD-1, may provide additional benefit by increasing im-mune checkpoint blockade drug delivery to CD8+ T cells, thereby en-hancing activity of immunotherapy. Clinical trials combiningmacrophagetargeting therapeutics and immune checkpoint blockers are underway (42)and should answer this question. In support of thenotion thatmyeloid cellsinterface with aPD-1 immunotherapy, recently identified correlates ofaPD-1 response in tumors suggest that alterations inmacrophage gene sig-natures are associated with nonresponsiveness to aPD-1 (43).

Comparisons between mouse and human Fc/Fc receptor inter-actions are complex because both antibody Fc isotypes and Fc receptorbiological functions are divergent between species. Our initial findingsin mice, which established the basic principle of Fc receptor–mediatedantibody transfer, were recapitulated using primary human cells andnivolumab. However, it will be important to understand the impactof such interactions in patients. In addition, one should clarify whetherall macrophages, or only a subset thereof, contribute to mAb uptake intumors. Our understanding of the complexity of macrophages and othermyeloid cells remains limited (18), and it is possible that those with dis-crete phenotypes, for example, with distinct proportions of activating ver-sus inhibitory Fc receptors, play different roles in response to immunecheckpoint blockade therapy (44). Uncovering themechanisms that con-trol mAb drug interactions with host cells is an ongoing challenge, al-though the findings could have far-reaching consequences for diagnosticand therapeutic approaches.

MATERIALS AND METHODSStudy designThe objective of this study was to understand the binding tropism ofaPD-1 mAb therapy in tumors. Flow cytometry, ex vivo microscopy,and intravital microscopy techniques were used for longitudinal inves-tigation of checkpoint blockade pharmacokinetics. The expectation wasthat the fluorescently tagged drug would allow tumor-infiltrating lym-

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

phocytes to be tracked during pharmacodynamic response; however,accumulation of drug in TAMs prompted us to hypothesize that the Fcregion of thesemAbs can also affect the drug biodistribution. All in vitroand in vivo studies were performed using C57BL/6J mice, and humanprimary cells were collected from healthy volunteers for ex vivo studies.Sample sizeswere decidedusing data frompreliminary calipermeasure-ments of tumor growth (SD = 25% with a 99% difference between treat-ment andcontrol groups;a =0.05,b =0.2).Datawere analyzedbyGrubbs’test for statistical outliers, which were predefined using an a value of 0.01.Data are representative of at least three independent experiments, as in-dicated in the figure legends. Treatment cohorts were assigned to ensurethat tumorswere size-matched at the start of the intervention.Calipermea-surements were acquired by researchers blinded to the intervention andgroupallocation.Microscopy imagingwasperformedunblindedusingpre-determined data collection methods that allowed multiple regions to bestudied for each sample, permitting thousands of cells to be analyzed.

Animal modelsAnimal research was performed in accordance with the InstitutionalAnimal Care and Use Committees at Massachusetts General Hospital(MGH). DPE-GFP mice (21) were provided by U. von Andrian (De-partment of Microbiology and Immunobiology, Harvard MedicalSchool). IFN-g reporter with endogenous polyadenylated tail (GREAT)mice (22) were provided byA. Luster (Division of Rheumatology, Allergyand Immunology,MGH). Tumor growthwasmonitored by calipermea-surement, and the area (A) of these predominantly two-dimensionaltumors was calculated using the formula A = length × width. Tumorimplantation was performed by intradermal injection of tumor cells(2 × 106 MC38/MC38–H2B-mApple, 5 × 105 B16-F1–ovalbumin, and5 × 105 KP1.9). MC38 cells were gifted by M. Smyth (QIMR Berghofer,Brisbane, Australia), B16 cells were fromAmerican Type Culture Collec-tion, and KP1.9 cells were a gift from A. Zippelius (University HospitalBasel, Switzerland). Experiments were generally started when tumorsbecame vascularized, which was after 8 days. For aPD-1 and AF647–aPD-1 treatments, mice were given 200 mg (intraperitoneally) of the29F.1A12 aPD-1 clone. For in vivo Fc blocking experiments, mice wereinfused intraperitoneally with 200 mg of monoclonal antibody specific tomouse FcgRII andFcgRIII (clone 2.4G2, Bioxcell) daily for 5days.Controlmice received 200 mg of rat IgG2a isotype control (clone 2A3, Bioxcell).

In vivo microscopyIntravital microscopy was performed in dorsal skinfold windowchambers installed on DPE-GFP or GREAT mice inoculated withMC38–H2B-mApple tumors. Mouse macrophages and/or vasculaturewere labeled with Pacific Blue–ferumoxytol and Pacific Blue–dextran,respectively. AF647–aPD-1 (200 mg) was delivered intravenously, andits tumor distribution was observed using an Olympus FluoViewFV1000MPE confocal imaging system (Olympus America), as de-scribed previously (45). Pacific Blue, GFP/YFP, mApple, and AF647were imaged sequentially using 405-, 473-, 559-, and 635-nm lasersand BA430-455, BA490-540, BA575-620, and BA575-675 emissionfilters with DM473, SDM560, and SDM640 beam splitters, all sourcedfrom Olympus America. Time-lapse images were acquired continuallyover the first hour after AF647–aPD-1 injection, after which the micewere allowed to recover before subsequent imaging.

Statistical analysisData points were compiled in Microsoft Excel, and statistical analyseswere performed using GraphPad Prism 6. a levels of 0.05 were used to

8 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

define statistical significance, and error bars represent SEM unless other-wise noted.

http://stm.scien

Dow

nloaded from

SUPPLEMENTARY MATERIALSwww.sciencetranslationalmedicine.org/cgi/content/full/9/389/eaal3604/DC1Materials and MethodsFig. S1. AF647–aPD-1 mAb quantification in various tissues at 24 hours after injection.Fig. S2. Specificity of T cell reporter mice and dextran nanoparticles.Fig. S3. Representative intravital microscopy time course of AF647–aPD-1 from an IFN-g–YFPmouse.Fig. S4. T cell and macrophage motility before and after aPD-1 treatment.Fig. S5. Assessment of aPD-1 binding to tumor cells.Fig. S6. aPD-1 transfer to tumor macrophages in vivo.Fig. S7. Distribution of AF647–aPD-1 across tumor models.Fig. S8. PD-1 expression by CD8+ T cells in the MC38 tumor microenvironment.Fig. S9. aPD-1 mAb transfer from T cells to macrophages in vitro.Fig. S10. Assessment of aPD-1 internalization after binding to PD-1.Fig. S11. aPD-1 degradation by macrophages.Fig. S12. PD-1 localization following aPD-1 internalization after binding to PD-1.Fig. S13. Comparative analysis of mAb glycosylation patterns between mouse and humanPD-1 antibodies.Fig. S14. Assessment of cell membrane component exchange during aPD-1 transfer.Fig. S15. Confirmation of deglycosylation and antigen binding affinity for rat anti-mouse PD-1.Fig. S16. Impact of Fc blockade on aPD-1 treatment efficacy.Fig. S17. Proposed resistance mechanism and potential improvement of aPD-1 mAb therapy.Movie S1. Intravital microscopy imaging of AF647–aPD-1 injection in a DPE-GFP mousebearing an MC38–H2B-mApple tumor in a dorsal skinfold chamber.Movie S2. Intravital microscopy imaging of AF647–aPD-1 injection in an IFN-g–enhanced YFPGREAT mouse bearing an MC38–H2B-mApple tumor in a dorsal skinfold chamber.Movie S3. Live imaging of mouse T cell aPD-1 transfer to macrophages.Movie S4. Live imaging of human T cell aPD-1 transfer to macrophages.References (46–52)

by guest on August 27, 2020

cemag.org/

REFERENCES AND NOTES1. D. R. Leach, M. F. Krummel, J. P. Allison, Enhancement of antitumor immunity by CTLA-4

blockade. Science 271, 1734–1736 (1996).2. D. M. Pardoll, The blockade of immune checkpoints in cancer immunotherapy. Nat. Rev.

Cancer 12, 252–264 (2012).3. S. L. Topalian, C. G. Drake, D. M. Pardoll, Immune checkpoint blockade: A common

denominator approach to cancer therapy. Cancer Cell 27, 450–461 (2015).4. K. M. Mahoney, P. D. Rennert, G. J. Freeman, Combination cancer immunotherapy and

new immunomodulatory targets. Nat. Rev. Drug Discov. 14, 561–584 (2015).5. G. P. Dunn, L. J. Old, R. D. Schreiber, The three Es of cancer immunoediting. Annu. Rev.

Immunol. 22, 329–360 (2004).6. R. D. Schreiber, L. J. Old, M. J. Smyth, Cancer immunoediting: Integrating immunity’s roles

in cancer suppression and promotion. Science 331, 1565–1570 (2011).7. D. S. Chen, I. Mellman, Oncology meets immunology: The cancer-immunity cycle.

Immunity 39, 1–10 (2013).8. S. L. Topalian, F. S. Hodi, J. R. Brahmer, S. N. Gettinger, D. C. Smith,

D. F. McDermott, J. D. Powderly, R. D. Carvajal, J. A. Sosman, M. B. Atkins,P. D. Leming, D. R. Spigel, S. J. Antonia, L. Horn, C. G. Drake, D. M. Pardoll, L. Chen,W. H. Sharfman, R. A. Anders, J. M. Taube, T. L. McMiller, H. Xu, A. J. Korman,M. Jure-Kunkel, S. Agrawal, D. McDonald, G. D. Kollia, A. Gupta, J. M. Wigginton,M. Sznol, Safety, activity, and immune correlates of anti–PD-1 antibody in cancer.N. Engl. J. Med. 366, 2443–2454 (2012).

9. E. B. Garon, N. A. Rizvi, R. Hui, N. Leighl, A. S. Balmanoukian, J. P. Eder, A. Patnaik,C. Aggarwal, M. Gubens, L. Horn, E. Carcereny, M.-J. Ahn, E. Felip, J.-S. Lee, M. D. Hellmann,O. Hamid, J. W. Goldman, J.-C. Soria, M. Dolled-Filhart, R. Z. Rutledge, J. Zhang,J. K. Lunceford, R. Rangwala, G. M. Lubiniecki, C. Roach, K. Emancipator, L. Gandhi;KEYNOTE-001 Investigators, Pembrolizumab for the treatment of non–small-cell lungcancer. N. Engl. J. Med. 372, 2018–2028 (2015).

10. P. C. Tumeh, C. L. Harview, J. H. Yearley, I. P. Shintaku, E. J. M. Taylor, L. Robert,B. Chmielowski, M. Spasic, G. Henry, V. Ciobanu, A. N. West, M. Carmona, C. Kivork, E. Seja,G. Cherry, A. J. Gutierrez, T. R. Grogan, C. Mateus, G. Tomasic, J. A. Glaspy, R. O. Emerson,H. Robins, R. H. Pierce, D. A. Elashoff, C. Robert, A. Ribas, PD-1 blockade induces responsesby inhibiting adaptive immune resistance. Nature 515, 568–571 (2014).

11. N. A. Rizvi, M. D. Hellmann, A. Snyder, P. Kvistborg, V. Makarov, J. J. Havel, W. Lee, J. Yuan,P. Wong, T. S. Ho, M. L. Miller, N. Rekhtman, A. L. Moreira, F. Ibrahim, C. Bruggeman,B. Gasmi, R. Zappasodi, Y. Maeda, C. Sander, E. B. Garon, T. Merghoub, J. D. Wolchok,

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

T. N. Schumacher, T. A. Chan, Mutational landscape determines sensitivity to PD-1 blockade in non–small cell lung cancer. Science 348, 124–128 (2015).

12. T. N. Schumacher, R. D. Schreiber, Neoantigens in cancer immunotherapy. Science 348,69–74 (2015).

13. M. S. Rooney, S. A. Shukla, C. J. Wu, G. Getz, N. Hacohen, Molecular and genetic propertiesof tumors associated with local immune cytolytic activity. Cell 160, 48–61 (2015).

14. C. Pfirschke, C. Engblom, S. Rickelt, V. Cortez-Retamozo, C. Garris, F. Pucci, T. Yamazaki,V. Poirier-Colame, A. Newton, Y. Redouane, Y.-J. Lin, G. Wojtkiewicz, Y. Iwamoto,M. Mino-Kenudson, T. G. Huynh, R. O. Hynes, G. J. Freeman, G. Kroemer, L. Zitvogel,R. Weissleder, M. J. Pittet, Immunogenic chemotherapy sensitizes tumors to checkpointblockade therapy. Immunity 44, 343–354 (2016).

15. A. J. Minn, E. J. Wherry, Combination cancer therapies with immune checkpoint blockade:Convergence on interferon signaling. Cell 165, 272–275 (2016).

16. T. F. Gajewski, H. Schreiber, Y.-X. Fu, Innate and adaptive immune cells in the tumormicroenvironment. Nat. Immunol. 14, 1014–1022 (2013).

17. R. Noy, J. W. Pollard, Tumor-associated macrophages: From mechanisms to therapy.Immunity 41, 49–61 (2014).

18. C. Engblom, C. Pfirschke, M. J. Pittet, The role of myeloid cells in cancer therapies.Nat. Rev. Cancer 16, 447–462 (2016).

19. K. J. Oestreich, H. Yoon, R. Ahmed, J. M. Boss, NFATc1 regulates PD-1 expression uponT cell activation. J. Immunol. 181, 4832–4839 (2008).

20. S. Gilfillan, C. J. Chan, M. Cella, N. M. Haynes, A. S. Rapaport, K. S. Boles, D. M. Andrews,M. J. Smyth, M. Colonna, DNAM-1 promotes activation of cytotoxic lymphocytes bynonprofessional antigen-presenting cells and tumors. J. Exp. Med. 205, 2965–2973 (2008).

21. T. R. Mempel, M. J. Pittet, K. Khazaie, W. Weninger, R. Weissleder, H. von Boehmer,U. H. von Andrian, Regulatory T cells reversibly suppress cytotoxic T cell functionindependent of effector differentiation. Immunity 25, 129–141 (2006).

22. R. L. Reinhardt, H.-E. Liang, K. Bao, A. E. Price, M. Mohrs, B. L. Kelly, R. M. Locksley,A novel model for IFN-g–mediated autoinflammatory syndromes. J. Immunol.194, 2358–2368 (2015).

23. R. Weissleder, M. Nahrendorf, M. J. Pittet, Imaging macrophages with nanoparticles.Nat. Mater. 13, 125–138 (2014).

24. F. Nimmerjahn, J. V. Ravetch, Divergent immunoglobulin g subclass activity throughselective Fc receptor binding. Science 310, 1510–1512 (2005).

25. R. Dahan, E. Sega, J. Engelhardt, M. Selby, A. J. Korman, J. V. Ravetch, FcgRs modulatethe anti-tumor activity of antibodies targeting the PD-1/PD-L1 axis. Cancer Cell 28,285–295 (2015).

26. P. V. Beum, A. D. Kennedy, M. E. Williams, M. A. Lindorfer, R. P. Taylor, The shavingreaction: Rituximab/CD20 complexes are removed from mantle cell lymphoma andchronic lymphocytic leukemia cells by THP-1 monocytes. J. Immunol. 176, 2600–2609(2006).

27. M. J. Feige, S. Nath, S. R. Catharino, D. Weinfurtner, Structure of the murineunglycosylated IgG1 Fc fragment. J. Mol. Biol. 391, 599–608 (2009).

28. R. P. Taylor, M. A. Lindorfer, Fcg-receptor–mediated trogocytosis impactsmAb-based therapies: Historical precedence and recent developments. Blood 125,762–766 (2015).

29. P. Bruhns, B. Iannascoli, P. England, D. A. Mancardi, N. Fernandez, S. Jorieux, M. Daeron,Specificity and affinity of human Fcg receptors and their polymorphic variants for humanIgG subclasses. Blood 113, 3716–3725 (2009).

30. A. Lux, X. Yu, C. N. Scanlan, F. Nimmerjahn, Impact of immune complex size andglycosylation on IgG binding to human FcgRs. J. Immunol. 190, 4315–4323 (2013).

31. C. Wang, K. B. Thudium, M. Han, X.-T. Wang, H. Huang, D. Feingersh, C. Garcia, Y. Wu,M. Kuhne, M. Srinivasan, S. Singh, S. Wong, N. Garner, H. Leblanc, R. T. Bunch, D. Blanset,M. J. Selby, A. J. Korman, In vitro characterization of the anti-PD-1 antibody nivolumab,BMS-936558, and in vivo toxicology in non-human primates. Cancer Immunol. Res. 2,846–856 (2014).

32. G. Suntharalingam, M. R. Perry, S. Ward, S. J. Brett, A. Castello-Cortes, M. D. Brunner,N. Panoskaltsis, Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibodyTGN1412. N. Engl. J. Med. 355, 1018–1028 (2006).

33. K. Hussain, C. E. Hargreaves, A. Roghanian, R. J. Oldham, H. T. C. Chan, C. I. Mockridge,F. Chowdhury, B. Frendeus, K. S. Harper, J. C. Strefford, M. S. Cragg, M. J. Glennie,A. P. Williams, R. R. French, Upregulation of FcgRIIb on monocytes is necessary topromote the superagonist activity of TGN1412. Blood 125, 102–110 (2015).

34. V. Goede, K. Fischer, R. Busch, A. Engelke, B. Eichhorst, C. M. Wendtner, T. Chagorova,J. de la Serna, M.-S. Dilhuydy, T. Illmer, S. Opat, C. J. Owen, O. Samoylova, K.-A. Kreuzer,S. Stilgenbauer, H. Döhner, A. W. Langerak, M. Ritgen, M. Kneba, E. Asikanius,K. Humphrey, M. Wenger, M. Hallek, Obinutuzumab plus chlorambucil in patients withCLL and coexisting conditions. N. Engl. J. Med. 370, 1101–1110 (2014).

35. F. Sclafani, D. Gonzalez de Castro, D. Cunningham, S. Hulkki Wilson, C. Peckitt,J. Capdevila, B. Glimelius, S. Roselló Keränen, A. Wotherspoon, G. Brown, D. Tait,R. Begum, J. Thomas, J. Oates, I. Chau, FcgRIIa and FcgRIIIa polymorphisms and cetuximabbenefit in the microscopic disease. Clin. Cancer Res. 20, 4511–4519 (2014).

9 of 10

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on August 27,

http://stm.sciencem

ag.org/D

ownloaded from

36. O. Manches, G. Lui, L. Chaperot, R. Gressin, J.-P. Molens, M.-C. Jacob, J.-J. Sotto,D. Leroux, J.-C. Bensa, J. Plumas, In vitro mechanisms of action of rituximab on primarynon-Hodgkin lymphomas. Blood 101, 949–954 (2003).

37. L. Arnould, M. Gelly, F. Penault-Llorca, L. Benoit, F. Bonnetain, C. Migeon, V. Cabaret,V. Fermeaux, P. Bertheau, J. Garnier, J.-F. Jeannin, B. Coudert, Trastuzumab-basedtreatment of HER2-positive breast cancer: An antibody-dependent cellular cytotoxicitymechanism? Br. J. Cancer 94, 259–267 (2006).

38. R. A. Clynes, T. L. Towers, L. G. Presta, J. V. Ravetch, Inhibitory Fc receptors modulate invivo cytotoxicity against tumor targets. Nat. Med. 6, 443–446 (2000).

39. A. Natarajan, A. T. Mayer, L. Xu, R. E. Reeves, J. Gano, S. S. Gambhir, Novel radiotracerfor immunoPET imaging of PD-1 checkpoint expression on tumor infiltratinglymphocytes. Bioconjug. Chem. 26, 2062–2069 (2015).

40. M. Hettich, F. Braun, M. D. Bartholomä, R. Schirmbeck, G. Niedermann, High-resolutionPET imaging with therapeutic antibody-based PD-1/PD-L1 checkpoint tracers.Theranostics 6, 1629–1640 (2016).

41. T. Schlothauer, S. Herter, C. F. Koller, S. Grau-Richards, V. Steinhart, C. Spick,M. Kubbies, C. Klein, P. Umaña, E. Mössner, Novel human IgG1 and IgG4 Fc-engineeredantibodies with completely abolished immune effector functions. Protein Eng. Des. Sel.29, 457–466 (2016).

42. P. Sharma, S. Hu-Lieskovan, J. A. Wargo, A. Ribas, Primary, adaptive, and acquiredresistance to cancer immunotherapy. Cancer Cell 168, 707–723 (2017).

43. W. Hugo, J. M. Zaretsky, L. Sun, C. Song, B. H. Moreno, S. Hu-Lieskovan, B. Berent-Maoz,J. Pang, B. Chmielowski, G. Cherry, E. Seja, S. Lomeli, X. Kong, M. C. Kelley, J. A. Sosman,D. B. Johnson, A. Ribas, R. S. Lo, Genomic and transcriptomic features of response to anti-PD-1 therapy in metastatic melanoma. Cell 165, 35–44 (2016).

44. B. Lehmann, M. Biburger, C. Brückner, A. Ipsen-Escobedo, S. Gordan, C. Lehmann,D. Voehringer, T. Winkler, N. Schaft, D. Dudziak, H. Sirbu, G. F. Weber, F. Nimmerjahn,Tumor location determines tissue-specific recruitment of tumor-associated macrophagesand antibody-dependent immunotherapy response. Sci. Immunol. 2, eaah6413 (2017).

45. G. M. Thurber, K. S. Yang, T. Reiner, R. H. Kohler, P. Sorger, T. Mitchison, R. Weissleder,Single-cell and subcellular pharmacokinetic imaging allows insight into drug actionin vivo. Nat. Commun. 4, 1504 (2013).

46. G. R. Guile, P. M. Rudd, D. R. Wing, S. B. Prime, R. A. Dwek, A rapid high-resolution high-performance liquid chromatographic method for separating glycan mixtures andanalyzing oligosaccharide profiles. Anal. Biochem. 240, 210–226 (1996).

47. R. J. Kurlander, J. Hall, Comparison of intravenous gamma globulin and a monoclonalanti-Fc receptor antibody as inhibitors of immune clearance in vivo in mice. J. Clin. Invest.77, 2010–2018 (1986).

48. K. S. Yang, R. H. Kohler, M. Landon, R. Giedt, R. Weissleder, Single cell resolution in vivoimaging of DNA damage following PARP inhibition. Sci. Rep. 5, 10129 (2015).

49. P. V. Beum, E. M. Peek, M. A. Lindorfer, F. J. Beurskens, P. J. Engelberts, P. W. H. I. Parren,J. G. J. van de Winkel, R. P. Taylor, Loss of CD20 and bound CD20 antibody fromopsonized B cells occurs more rapidly because of trogocytosis mediated by Fc receptor-expressing effector cells than direct internalization by the B cells. J. Immunol. 187,3438–3447 (2011).

50. M. Alieva, L. Ritsma, R. J. Giedt, R. Weissleder, Imaging windows for long-termintravital imaging: General overview and technical insights. IntraVital 3,e29917 (2014).

Arlauckas et al., Sci. Transl. Med. 9, eaal3604 (2017) 10 May 2017

51. T. R. Mempel, S. E. Henrickson, U. H. von Andrian, T-cell priming by dendritic cells inlymph nodes occurs in three distinct phases. Nature 427, 154–159 (2004).

52. F. Nimmerjahn, P. Bruhns, K. Horiuchi, J. V. Ravetch, FcgRIV: A novel FcR with distinct IgGsubclass specificity. Immunity 23, 41–51 (2005).

Acknowledgments: We thank A. Luster and U. von Andrian for mice; M. Smyth andA. Zippelius for cell lines; T. Mempel, C. Engblom, and C. Pfirschke for helpful discussions;and G. Wojtkiewicz and B. Tricot for technical assistance. Funding: This work was supportedin part by the Samana Cay MGH Research Scholar Fund, the NIH (grants R01AI084880,R01CA164448, R21CA190344, U54-CA151884, 5P50CA086355, DP2AR068272-01, andHL084312), U.S. Department of Defense (PC140318), and the David H. Koch–Prostate CancerFoundation Award in Nanotherapeutics. G.J.F. was supported by P50CA101942. S.P.A.,M.F.C., and M.A.M. were supported by T32 CA79443. C.S.G. was supported by F31 CA196035.Author contributions: S.P.A., C.S.G., R.W., and M.J.P. developed the concept. S.P.A., R.W.,and M.J.P. developed intravital imaging studies. S.P.A., C.S.G., and M.J.P. designed theexperiments. S.P.A, C.S.G., R.H.K., M.F.C., K.S.Y., M.K., M.A.M., and J.C.C. performed theexperiments. S.P.A., C.S.G., R.W., and M.J.P. wrote the paper. All authors analyzed the resultsand edited the manuscript. Competing interests: G.J.F. is an inventor on U.S. patents6808710, 7101550, 7105328, 7638492, 7700301, 7432059, 7709214, 7722868, 7635757, and7038013 held by the Dana-Farber Cancer Institute that cover aspects of PD-1, PD-L1, andPD-L2 immunotherapy. G.J.F. has patents/pending royalties on the PD-1 pathway from Roche,Merck, Bristol-Myers Squibb, EMD Serono, Boehringer Ingelheim, AstraZeneca, and Novartis.G.J.F. has served on advisory boards for CoStim, Novartis, Roche, Eli Lilly, Bristol-Myers Squibb,Seattle Genetics, Bethyl Laboratories, Xios, and Quiet. M.J.P. has served as a consultant forBaxalta, Deciphera Pharmaceuticals, FORMA Therapeutics, Incyte Pharmaceuticals, JounceTherapeutics, KSQ Therapeutics, Secarna, Siamab, and Syndax; these commercial relationshipsare unrelated to the current study. R.M.A. is an inventor on U.S. patents 7846744 held bythe Rockefeller University and a provisional patent WO2016144760 held by MGH andon the scientific advisory board of Pionyr Immunotherapeutics. R.W. is a cofounder of T2Biosystems, is a cofounder and on the Board of Directors of Lumicell, and serves as a scientificadviser for Moderna Therapeutics, Tarveda Therapeutics, and Alivio Therapeutics. Theother authors declare that they have no competing interests. Data and materials availability:All cell lines were obtained through material transfer agreements or are commerciallyavailable. Requests for collaboration involving materials used in this research will befulfilled provided that a written agreement is executed in advance between MGH and therequesting parties.

Submitted 1 October 2016Resubmitted 7 November 2016Accepted 16 March 2017Published 10 May 201710.1126/scitranslmed.aal3604

Citation: S. P. Arlauckas, C. S. Garris, R. H. Kohler, M. Kitaoka, M. F. Cuccarese, K. S. Yang,M. A. Miller, J. C. Carlson, G. J. Freeman, R. M. Anthony, R. Weissleder, M. J. Pittet, In vivoimaging reveals a tumor-associated macrophage–mediated resistance pathway in anti–PD-1 therapy. Sci. Transl. Med. 9, eaal3604 (2017).

202

10 of 10

0

PD-1 therapy−antimediated resistance pathway in−In vivo imaging reveals a tumor-associated macrophage

Miles A. Miller, Jonathan C. Carlson, Gordon J. Freeman, Robert M. Anthony, Ralph Weissleder and Mikael J. PittetSean P. Arlauckas, Christopher S. Garris, Rainer H. Kohler, Maya Kitaoka, Michael F. Cuccarese, Katherine S. Yang,

DOI: 10.1126/scitranslmed.aal3604, eaal3604.9Sci Transl Med

vivo.PD-1 and prolonged its effects in− receptors prevented removal of antiγproblem, showing that inhibition of Fc

antibodies from T cells, thus inactivating them. The researchers also identified a potential way to overcome thisbound to T cells as intended, the authors found that tumor-associated macrophages quickly removed these

PD-1 antibodies initially−. discovered. Although antiet alUnfortunately, this does not always happen, as Arlauckas PD-1 antibodies must be able to remain bound to the T cells.−preventing T cells from becoming exhausted, anti

prominence in cancer treatment, but even these promising therapeutics do not always work. To be effective in 1 (PD-1) are gaining increasing−Antibodies against immune checkpoints such as programmed death

PD-1−Tug-of-war with anti

ARTICLE TOOLS http://stm.sciencemag.org/content/9/389/eaal3604

MATERIALSSUPPLEMENTARY http://stm.sciencemag.org/content/suppl/2017/05/08/9.389.eaal3604.DC1

CONTENTRELATED

http://science.sciencemag.org/content/sci/368/6487/181.fullhttp://stm.sciencemag.org/content/scitransmed/12/530/eaax6337.fullhttp://science.sciencemag.org/content/sci/364/6440/558.fullhttp://stm.sciencemag.org/content/scitransmed/11/477/eaat7973.fullhttp://science.sciencemag.org/content/sci/362/6411/eaar3593.fullhttp://stm.sciencemag.org/content/scitransmed/10/436/eaan3311.fullhttp://stm.sciencemag.org/content/scitransmed/9/415/eaan0401.fullhttp://science.sciencemag.org/content/sci/357/6349/409.fullhttp://science.sciencemag.org/content/sci/357/6349/358.fullhttp://stm.sciencemag.org/content/scitransmed/8/370/370ra180.fullhttp://stm.sciencemag.org/content/scitransmed/9/379/eaah3560.fullhttp://stm.sciencemag.org/content/scitransmed/9/385/eaak9679.fullhttp://stm.sciencemag.org/content/scitransmed/9/385/eaak9670.full

REFERENCES

http://stm.sciencemag.org/content/9/389/eaal3604#BIBLThis article cites 52 articles, 18 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

registered trademark of AAAS. is aScience Translational MedicineScience, 1200 New York Avenue NW, Washington, DC 20005. The title

(ISSN 1946-6242) is published by the American Association for the Advancement ofScience Translational Medicine

Copyright © 2017, American Association for the Advancement of Science

by guest on August 27, 2020

http://stm.sciencem

ag.org/D

ownloaded from